Monitoring And Diagnosis Of Oscillation In Control Loops

In the process industries, excessively oscillatory feedback control loops are a common occurrence due to poor controller tuning, control valve stiction, poor process and control system design, and oscillatory disturbances.The presence of oscillations in a plant increases the variability of the process variables and thus may cause poor control performance, inferior quality products, larger rejection rates and less plant profitability. It is clearly desirable to be able to detect excessively oscillatory loops, identify the source of the oscillations, and if possible, eliminate it. So the method for monitoring and diagnosing of plant-wide oscillations is given to detect oscillations and find the source.In this paper, the method for monitoring oscillations has been improved, and the method for diagnosing oscillations has been combined, for the seek of establishing a set of monitoring and diagnosis strategy with good industrial practicability, rapidity and accuracy. Main research work includes as follows:(1) Oscillation monitoring : monitoring the oscillations has got some achievement, however , there is still room for improvement in many respects such as the anti-noise, quantification, simplify, and so on. Based on the frequency domain method of monitoring, an oscillation index is proposed , and its anti-noise capability is estimated. In addition, the reference value for the significant threshold is given for the further distinction of oscillation significant level. Finally, the flow for monitoring the oscillations of industrial process is proposed based on the oscillation index and significant threshold.(2) Oscillation diagnosis: the course of diagnosis may be divided into two stages, In the first stage, the loops which are nearest to the sources need to be found, and the sources need to be identified in those loops in the second stage. Acorrding to this flow, various methods for classification and diagnosis of osicllations are compared and combined, therefore a set of oscillation strategy for industrial process diagnosis is formed. The simulation results show that the application of this flow can effectively reduce the range of the oscillation sources.(3) Industrial application : a flow is formed using those methods which are proposed or concluded to monitor and diagnose the oscillations of PTA production plant in Yangzi Petrochemical Corporation. There are some practical problems in the process of monitoring and diagnosis, to solve them the necessary analysis and explanation are given in details. The results indicate that using the flow is helpful to the detection of oscillations of plant in time and positioning of the sources accurately. What's more, it possesses a certain degree of stability for different sets of data. Those advantages make it very suitable for practical industrial application.